GR.7 Artificial intelligence-based decision support predicts requirement for neurosurgical intervention in acute traumatic brain injury
Bibliographic record
Abstract
Background: We aimed to develop an efficient and reliable artificial intelligence solution to automate prediction of neurosurgical intervention using acute traumatic brain injury computed tomography (CT) scans. Methods: TBI patients were identified from 2005 - 2022 at a Level 1 Canadian trauma center. Model training, validation, and testing was performed using head CT scans with patient-level labels corresponding to whether the patient received neurosurgical intervention. The finalized model was then deployed in a simulated prospective fashion on all TBI patients presenting to our center over an 18-month epoch. Results: 2,806 TBI scans were utilized for development of the Automated Surgical Intervention Support Tool (ASIST-TBI). 612 additional consecutive scans were used for simulated prospective model deployment. Prediction of neurosurgical intervention exhibited an area under receiver operating curve (AUC) of 0.92, accuracy of 0.87, sensitivity of 0.87, and specificity of 0.88 on the test dataset. On simulated prospective data, the results were: AUC 0.89, sensitivity 0.85, specificity 0.84 and accuracy of 0.84. Conclusions: We demonstrate the development and validation of ASIST-TBI, a machine learning model that accurately predicts whether TBI patients will need neurosurgical intervention. This model has potential application to optimize decision support and province-wide efficiency of inter-facility TBI triage to tertiary care centers.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".